In this problem, we using a novel deep reinforcement learning technique to achieve state-of-the-art results in the Atari game: Kangaroo. Out current plan is to release the results of all Atari games in January 2019, but for now…We leave you with this preview :) [Read More]

In this episode, we use a prebuilt experiment in the project malmo python examples folder: Tutorial_6.py. We then modified the file to include the agent algorithm to solve this problem. In this experiment, Ralph solved the problem using Q-Learning. The idea behind Q-Learning is that with each state (which block... [Read More]

I was attempting to train a neural network model on a single GPU in a multi-GPU Linux machine. I followed the advice that others had posted online to add the following lines at the beginning of my python script: [Read More]